XB-ART-47673Development December 1, 2013; 140 (24): 4903-13.
The Xenopus homologue of Down syndrome critical region protein 6 drives dorsoanterior gene expression and embryonic axis formation by antagonising polycomb group proteins.
Mesoderm and embryonic axis formation in vertebrates is mediated by maternal and zygotic factors that activate the expression of target genes. Transcriptional derepression plays an important role in the regulation of expression in different contexts; however, its involvement and possible mechanism in mesoderm and embryonic axis formation are largely unknown. Here we demonstrate that XDSCR6, a Xenopus homologue of human Down syndrome critical region protein 6 (DSCR6, or RIPPLY3), regulates mesoderm and embryonic axis formation through derepression of polycomb group (PcG) proteins. Xdscr6 maternal mRNA is enriched in the endoderm of the early gastrula and potently triggers the formation of dorsal mesoderm and neural tissues in ectoderm explants; it also dorsalises ventral mesoderm during gastrulation and induces a secondary embryonic axis. A WRPW motif, which is present in all DSCR6 homologues, is necessary and sufficient for the dorsal mesoderm- and axis-inducing activity. Knockdown of Xdscr6 inhibits dorsal mesoderm gene expression and results in head deficiency. We further show that XDSCR6 physically interacts with PcG proteins through the WRPW motif, preventing the formation of PcG bodies and antagonising their repressor activity in embryonic axis formation. By chromatin immunoprecipitation, we demonstrate that XDSCR6 releases PcG proteins from chromatin and allows dorsal mesoderm gene transcription. Our studies suggest that XDSCR6 might function to sequester PcG proteins and identify a novel derepression mechanism implicated in embryonic induction and axis formation.
PubMed ID: 24301465
Article link: Development
Genes referenced: bmi1 chrd.1 egr2 ezh2 gsc lefty mlc1 muc2 myc nodal nodal1 nodal2 nodal5 nodal6 nog odc1 otx2 prc1 ripply2.1 ripply2.2 ripply3 sox3 tbxt tle4 vegt wnt8a
Article Images: [+] show captions
|Fig. 1. XDSCR6 triggers the formation of dorsal mesoderm and neural tissues in Xenopus ectoderm explants. (A,B) Morphology of control (A) and Xdscr6-injected (B) explants at the early neurula stage equivalent. (C-H) In situ hybridisation analyses of the expression of MLC, XCG1 and Sox3 genes in control (C,E,G) and Xdscr6-injected (D,F,H) ectoderm explants at the late neurula stage equivalent. (I) RT-PCR analyses of gene expression in ectoderm explants cultured to stage 10.5 and 20 equivalent, following injection of the indicated amounts of Xdscr6 mRNA, Xdscr6 DNA or after activin treatment. ODC was used as a loading control. Ms., muscle specific. (J) Real-time RT-PCR analyses of the expression level of the indicated genes in early gastrula ectoderm explants injected with low and high amounts of Xdscr6 mRNA (top row) or treated with low and high doses of activin (bottom row). Similar results were obtained using ODC or EF1α as reference.|
|Fig. 2. XDSCR6 induces a secondary embryonic axis. (A) Xdscr6 mRNA or DNA was injected into one ventral blastomere (arrow) at the 4-cell stage. (B,C) Vegetal view of an uninjected early gastrula embryo (B) and a Xdscr6 mRNA-injected early gastrula embryo showing the precocious formation of the ventral blastopore (C, arrow). (D,E) Dorsal view, with anterior region at the top, of an uninjected late neurula embryo (D) and a Xdscr6 mRNA-injected late neurula embryo with a secondary neural tube (E). (F-H) Phenotypes of control (F), Xdscr6 mRNA-injected (G) and Xdscr6 DNA-injected (H) embryos at stage 35. (I,J) β-galactosidase staining of stage 32 embryos injected with lacZ mRNA alone (I) or co-injected with Xdscr6 mRNA (J). (K) Transverse section corresponding to J showing β-galactosidase staining in XDSCR6-induced muscles. (L-O) In situ hybridisation analyses of the expression of the indicated genes in Xdscr6 mRNA-injected embryos at stage 32. I and II designate the primary axis and the secondary axis, respectively.|
|Fig. 3. Dorsalisation of ventral mesoderm by XDSCR6. (A-H) In situ hybridisation analyses of Otx2, chordin, Wnt8 and Xvent1 expression in uninjected (A,C,E,G) and Xdscr6-injected (B,D,F,H) embryos at the early gastrula stage. Ectopic Otx2 and chordin expression sites are indicated by arrowheads. (I) Scheme of the dorsalisation experiment. Embryos were ventrally injected at the 4-cell stage (double arrows) and DMZ (D) or VMZ (V) explants were dissected at the early gastrula stage and cultured to late neurula stage equivalent. (J-M) Morphology of the uninjected and injected DMZ or VMZ explants. (N) RT-PCR analyses of gene expression in uninjected and injected DMZ or VMZ explants. (O) In situ hybridisation of MLC and XCG1 expression in ectoderm explants injected with the mRNAs indicated at the top and cultured to late neurula stage equivalent. DMZ, dorsal marginal zone; VMZ, ventral marginal zone.|
|Fig. 4. XDSCR6 is required for anterior development. (A,B) Dorso-vegetal view of chordin expression in a CoMO (A) and a Xdscr6MO-injected (B) embryo at stage 11. (C,D) Dorso-vegetal view of noggin expression in a CoMO (C) and a Xdscr6MO-injected (D) embryo at stage 10.5. (E,F) Dorso-vegetal view of goosecoid expression in a CoMO (E) and a Xdscr6MO-injected (F) embryo at stage 10. (G,H) Anterior view of Otx2 expression in a CoMo (G) and a Xdscr6MO-injected (H) embryo at stage 20. (I,J) Ventral view, with anterior region at the top, of XCG1 expression in a CoMO (I) and a Xdscr6MO-injected (J) embryo at stage 20. (K,L) Dorsal view, with anterior region at the top, of Sox3 expression in a CoMO (K) and a Xdscr6MO-injected (L) embryo at stage 18. (M-O) Phenotypes of a CoMo-injected embryo (M), a Xdscr6 morphant embryo (N) and a myc-Xdscr6 rescued embryo (O) at stage 38. (P) Summary of the rescue results. Numbers at the top indicate total embryos scored from four independent experiments. (Q) Western blot analysis shows the blockade of protein synthesis from Xdscr6-Flag mRNA, but not from myc-Xdscr6 mRNA, by Xdscr6MO. Tubulin is a loading control. CoMO, mismatch control MO.|
|Fig. 5. The WRPW motif is necessary and sufficient to induce a secondary axis. (A) Structure of the various Xdscr6 constructs. (B) Summary of secondary axis formation in control (Ctl) and injected embryos (constructs numbered as in A). XDSCR6 (1), XDSCR6N (2), XDSCR6-EnR (5), XDSCR6-VP16 (6), myc-WRPW (7) and GFP-WRPW (9) similarly induce a secondary axis, whereas XDSCR6C (3), XDSCR6m (4), myc-WRSG (8) and GFP-WRSG (10) have no effect. The results are expressed as a percentage and numbers at the top indicate total embryos scored from three independent experiments. (C) A GFP-WRPW-injected stage 35 embryo with green fluorescence concentrated in the anterior region of the secondary axis (arrow). (D) A GFP-WRSG-injected embryo lacks a secondary axis, with green fluorescence distributed in the posterior region. (E) In situ hybridisation analyses of chordin expression at stage 10.5 following overexpression of XDSCR6-EnR, XDSCR6-VP16 or myc-WRPW in the ventro-vegetal region at the 4-cell stage. All embryos are in vegetal view.|
|Fig. 6. XDSCR6 physically interacts with PcG proteins. (A) GST pull-down of embryonically expressed XEzh2-myc or XGrg4-myc by GST-XDSCR6. (B) Co-immunoprecipitation of XEzh2-myc by XDSCR6-Flag. (C) GST pull-down of embryonically expressed XDSCR6-myc, but not the mutant XDSCR6m-myc, by GST-BMI1. (D) Co-immunoprecipitation of XDSCR6-myc by BMI1-Flag. (E) GST pull-down of in vitro translated XDSCR6-myc, but not the mutant XDSCR6m-myc, by GST-XEzh2 or GST-BMI1. (F) GST pull-down of in vitro translated BMI1-GFP or XEzh2-myc by GST-XDSCR6. IP, immunoprecipitation; WB, western blot.|
|Fig. 7. XDSCR6 interferes with the nuclear localisation of endogenous Ezh2 and BMI1. Confocal microscopy analyses of Ezh2- and BMI1-associated PcG bodies in HeLa cells. In cells transfected with Xdscr6-RFP or myc-WRPW (arrowheads in A-C,M-O,I-K,U-W), the intensity of endogenous Ezh2 and BMI1 staining is significantly decreased compared with neighbouring untransfected cells in the same culture dish (arrows). In Xdscr6m-RFP-transfected cells (arrowheads in E-G,Q-S), the intensity of fluorescence is comparable to that of neighbouring untransfected cells (arrows). Scale bar: 10 μm. (D,H,L,P,T,X) Quantification of nuclear fluorescence intensity. Red lines represent the median for each condition. Data are presented as mean ± s.e.m. P-values (Student’s t-test) and number of measures (n) are indicated.|
|Fig. 8. XDSCR6 releases endogenous Ezh2 and BMI1 from the goosecoid promoter. ChIP experiments using uninjected (uninj.) or Xdscr6-injected Xenopus ectoderm explants. (A) A representative semi-quantitative PCR result. (B-E) Quantitative PCR analyses of the goosecoid promoter using chromatin immunoprecipitated from uninjected or Xdscr6-injected ectoderm explants by the indicated antibodies. Data represent the mean of triplicate experiments (error bars indicate s.d.).|
|Fig. 9. XDSCR6 and BMI1 are mutually antagonistic during embryonic axis formation. (A-D) XDSCR6 rescues anterior deficiency produced by BMI1 overexpression. Embryos at the 4-cell stage were either left uninjected (A) or dorsally injected with Xdscr6 mRNA (B), BMI1 mRNA (C) or co-injected with BMI1 and Xdscr6 mRNAs (D), and cultured to the stage 32. (E-H) BMI1 inhibits the axis-inducing activity of XDSCR6. Embryos at the 4-cell stage were either left uninjected (E) or ventrally injected with Xdscr6 mRNA (F), BMI1 mRNA (G) or co-injected with BMI1 and Xdscr6 mRNAs (H), and cultured to the stage 35. (I,J) BMI1 represses the goosecoid promoter (gsc-luc) and Xnr1 promoter (Xnr1-luc) activity that had been stimulated by overexpression of XDSCR6 in ectoderm explants. Values were expressed relative to the value obtained from uninjected control explants. Mean ± s.e.m.; P<0.05 (Student’s t-test). RLU, relative luciferase units. (K) Model of XDSCR6 function in transcriptional derepression. (Top) In the absence of XDSCR6, PcG proteins bind to chromatin and repress the transcription of mesoderm genes. (Bottom) The interaction between XDSCR6 and PcG proteins removes PRC1 and PRC2 from chromatin and contributes to transcriptional derepression.|
|Fig. S1. RT-PCR analyses of the expression and distribution of Xdscr6 transcripts during early development. (A) Temporal expression of Xdscr6 at various stages (numbers on the top) and its distribution at stage 35 in the head (He), trunk (Tr) and posterior (P) regions. (B) Schema showing the dissection of different early gastrula tissues including ectoderm (ect), mesoderm, endoderm, dorsal endoderm (DE) and ventral endoderm (VE). (C) RT-PCR analyses of the distribution of Xdscr6 transcripts in different regions of the early gastrula. (D) Induction of Xdscr6 expression in early gastrula ectoderm following overexpression of squint and Wnt8. ODC was used as an input control.|
|Fig. S2. XDSCR6 differentially induces the expression of nodal-related genes. Embryos at the 2-cell stage were injected at the animal pole region with Xdscr6 mRNA (0.5 ng), RT-PCR analyses were performed on control embryo and dissected ectoderm explants at the early gastrula stage.|
|Fig. S3. Comparison of sequence and activity between Xenopus and human RIPPLY2 proteins. (A) Sequence alignment. Conserved residues are indicated in red. The amino- and carboxyl-terminal halves of XDSCR6 used for functional analyses are indicated. (B) RT-PCR analyses of the activity of XDSCR6 and human RIPPLY2 in mesoderm gene expression at the early gastrula stage. (C) A control embryo at the stage 33. (D) Induction of a partial secondary axis by human RIPPLY2. The frequency of this induction is similar as XDSCR6 (not shown).|
|Fig. S4. Uniform expression of XDSCR6 suppresses trunk and posterior development. Embryos at the 8-cell stage were radially injected with Xdscr6 mRNA (0.4 ng) at the equatorial region and live images of control (A) and injected (B) embryos were taken at larval stage 41. Note that embryos radially injected with Xdscr6 lacks trunk and posterior regions.|
|Fig. S5. XDSCR6 induces chordin and Xbra expression independently of nodal signalling. Embryos at the 2-cell stage were injected at the animal pole region with low amount of Xdscr6 mRNA alone or coinjected with VegT-EnR mRNA or antivin mRNA. They were also injected with VegT mRNA (40 pg) or coinjected with VegT-EnR mRNA or antivin mRNA. RT-PCR analyses on control embryo and dissected ectoderm explants were performed at the early gastrula stage. VegT-EnR and antivin does not block the activity of XDSCR6, but inhibits the activity of VegT.|
|Fig. S6. RT-PCR analyses of Otx2, chordin and Xbra expression in ectoderm explants injected with wild-type Xdscr6 or its mutants, as indicated.|
|Fig. S7. BMI1 inhibits ectopic chordin expression induced by XDSCR6. Embryos at the 4-cell stage were ventrally injected with Xdscr6 mRNA alone, BMI1 mRNA alone or coinjected with Xdscr6 and BMI1 mRNA. In situ hybridization was performed at the early gastrula stage.|